This proposal outlines analyses to be done by the St. Louis site of data collected by 2 interviews a year apart face-to-face and 1 telephone interview at the midpoint between the interviews. Analyses are divided between cross-site and individual site. Cross-site analyses, in which St. Louis would take a leadership role in collaborating with the other 4 ECA sites, will cover the diagnoses of antisocial personality and alcohol abuse and dependence, and studies of prisoners, the general population of young adults, black-white differences, and fecundity. Individual analyses will involve selection of symptoms for an efficient screening interview, the validity of each DSM-III diagnosis in the DIS in a general population sample, a test of Brown's hypothesis about factors leading to depression in urban women, the effects of forgetting on report of lifetime symptoms, the frequency of recovery to normal drinking in persons with a history of alcohol dependence, the natural history of alcohol sysmptoms, the relation of psychiatric diagnosis to 1-year mortality, and the prevalence and incidence of diagnoses available only in the St. Louis site. Plans are also described for a study of the effect of neighborhood change on psychiatric disorder, studies of social supports and daily functioning as related to diagnosis, and methodological studies of the effects of degree of effort to get a high response rate on estimates of prevalence.